Although electroconvulsive therapy (ECT) is recognized to be highly effective and is extensively used in the management of severe depression, an optimum treatment schedule has never been definitively established. Clinical practice continues to vary widely in terms of frequency of administration and number of ECTs in the series. Interim results from the present project confirm that these parameters can substantially influence both rate of antidepressant response and cognitive adverse effects. Compared to twice weekly treatment (ECTx2), patients who received ECTx3 per week responded significantly more rapidly but were also more impaired on neuropsychological testing at the end of the four week trial (although not at 1 and 6 month follow-up). The next stage of the project will focus on optimization of the more rapidly effective ECTx3 schedule by reducing the number of ECT's administered to this group. This strategy is justified by the previous findings and is expected to significantly diminish cognitive morbidity without altering rate of response or final outcome. Double-blind comparison with ECTx2 will be achieved by the use of simulated ECT and the groups will be balanced in terms of number of real (n=8) and simulated (n=4) treatments administered during the 4 week acute phase of the study. Long term effects will again be assessed during 6 month follow-up under lithium carbonate continuation therapy. A central objective of the project has been to correlate biological effects of ECT with optimally effective treatment. The next phase of the study will further characterize the significant enhancement by ECT of serotonergically mediated hormone release which was previously demonstrated. The time course and long term status of this effect will be examined and specific neurochemical components identified by the introduction of an additional pharmaco-endocrine (fenfluramine or ipsapirone) challenge paradigm.